Most genetic diseases that affect the central nervous system (CNS) produce metabolic abnormalities and lesions throughout the brain. Somatic gene transfer is potentially capable of arresting or reversing the disease process in many genetic disorders. However, the methods presently available to transfer genes into the central nervous system generally only deliver a foreign gene to a limited number of cells or to specific structures. A large class of human genetic diseases affecting the CNS are the lysosomal storage disorders, which are characterized by mental retardation and other progressive degenerative abnormalities. The long-term goal of our studies is to develop a method to deliver normal lysosomal enzyme genes into the brain in a manner that can correct the disseminated lesions that are typical of this group of neurodegenerative diseases. In this grant we will study the properties of adeno-associated virus (AAV) vector-mediated transfer of a therapeutic gene, beta-glucuronidase (GUSB), in the brains of GUSB-deficient animal models of mucopolysaccharidosis (MPS) type VII (Sly disease), a representative neurodegenerative storage disease. Preliminary studies demonstrate that the vectors can express sustained therapeutic levels of the transferred GUSB, and secrete the enzyme in a large sphere of correction surrounding the transduction site, suggesting that this vector may be suitable for long-term correction of genetic deficiencies. The Specific Aims are to: 1) Determine the effects on vector gene expression promoters that are expressed widely in the brain or at high levels in neurons; 2) Characterize the expression of GUSB from AAV vectors in various cell types and structures of the brain; 3) Quantitate the amount and longevity of expression of GUSB in the MPS VII brain and evaluate methods which may increase total transduction and enzyme delivery; and 4) Evaluate changes in pathology in the brain over the long-term. Most of the development and testing will be done in the MPS VII mouse brain to optimize the vector designs and delivery strategies. Then, the methods will be tested in the MPS VII dog brain to evaluate vector performance and enzyme diffusion in a larger volume of brain tissue. Evaluation in true clinical models of a human neurodegenerative disease will help define both the potential and limitations of this approach to treating the CNS disease in lysosomal storage disorders.